ICON 2018ICON 2018 - Tutorial

Advancements in Computational Psycholinguistics: Approach, Experiment and Inference

Abstract

Language as a communication tool is one of the key attributes of human society. It also distinguishes human communication from most of the other species. However, language is a complex and intricate tool developed and evolved over thousands of years, influenced by usage, demographics, and socio-cultural factors. The study of language communication and comprehension is a rapidly expanding and multi-disciplinary and challenging field of research.
Due to technological advances in computational methods that capture the statistical and formal properties, significant progress has been made possible in the understanding of how human process language. One of such approaches is the paradigm of Computational Psycholinguistics. Computational Psycholinguistics applies state of the art machine learning techniques to simulate the relationship between human behavior, psychological process and language comprehension and develops prediction models. Computational Psycholinguistics is not only an important paradigm for gaining better insights into how ‘brain raeds words’, rather it has been observed that a number of NLP and IR tasks such as the design of a knowledge base, enhancing textbooks for target readers can be improved based on the understanding of the process of human language comprehension. Hence, computational psycholinguistics is useful tool for researchers across multiple disciplines such as linguists, technical researchers working in NLP, cognitive scientists etc.

Duration

The tutorial will start with a recap of Computational Psycholinguistics , its role in the study of language comprehension. Then the tutorial will focus on development and design of psycholinguistics to study human language comprehension and some example case studies.

The outline of the tutorial topics is as follows:

  1. Introduction to Computational Psycholinguistics
    1. Aims and objective of psycholinguistics studies
    2. Methodologies of psycholinguistics studies
    3. Design and development of psycholinguistics experiments
  2. The Mental Lexicon
    1. Organization and Processing of words
    2. Current theories of Mental Lexicon (ML) and related key questions
    3. Experimental techniques for probing ML
    4. Computational models of ML
    5. Applications of ML models in NLP
    6. Applications of NLP and Machine learning techniques in ML research
  1. Processing of morphologically complex words in the mental lexicon
    1. Inflections
    2. Derivations
    3. Compounding
    4. Complex predicates
  2. Compositionality and Computational Models
  3. Role of Orthography and Phonology in Word Recognition
  4. Case studies and Experiments
    1. Discussions and demonstrations some representative psycholinguistics studies to infer on human language comprehension
  5. Q&A

Manjira Sinha

Manjira Sinha is Assistant Professor, Centre for Educational technology, Indian Institute of Technology Kharagpur. Prior to that she was a research scientist in the Text and Graph Analytics research group at Conduent Labs India (formerly known as Xerox Research Centre India) for 3 years. She is currently working in NLP for Healthcare. She has also worked on cross-domain Text categorization, Social Media analysis for Urban Informatics, Knowledge Extraction, and Quality Analysis for Call center Interactions. Manjira has a Ph.D. in Computer Science from the Indian Institute of Technology Kharagpur. She is also visiting faculty in Indian Institute of Information Technology Kalyani. Her areas of interest include Language Comprehension and Psycholinguistics, Natural Language Processing, Assistive Technology and Human Computer Interaction. She has organized a number of workshops and tutorials including the 1st workshop on Language, Cognition and Computational Models in Coling 2018.

Linkedin Profile

Tirthankar Dasgupta

Tirthankar Dasgupta is a research scientist at Innovation Labs, Tata Consultancy Services Ltd., India in the Text Analytics and Web Intelligence group. He holds a Ph.D. in computer science from Indian Institute of Technology, Kharagpur. His research interests span natural language processing, computational psycholinguistics, machine learning and human computer interaction. He has organized a number of workshops in the area of assistive technology and natural language processing. He is also an active organizing member and regional coordinator of Panini Linguistic Olympiad in India. He was an organizing member of International Linguistic Olympiad 2016 in India. Tithankar is the co-organizer of the 1st workshop on Language, Cognition and Computational Models in Coling 2018.

Linkedin Profile